Enhancing signature word detection in voice assistants
AssigneeAdeia Guides Inc.
InventorsAnkur Anil Aher, Jeffry Copps Robert Jose
Life of this patent
Estimating the term…
Abstract
Systems and methods detecting a spoken sentence in a speech recognition system are disclosed herein. Speech data is buffered based on an audio signal captured at a computing device operating in an active mode. The speech data is buffered irrespective of whether the speech data comprises a signature word. The buffered speech data is processed to detect a presence of the sentence comprising at least one command and a query for the computing device. Processing the buffered speech data includes detecting the signature word in the buffered speech data, and in response to detecting the signature word in the speech data, initiating detection of the sentence in the buffered speech data.
Claims19
A computer-implemented method comprising:
- detecting, while a speech recognition device is in a non-active mode and a signature word is not detected to enable an active mode, a user activity that indicates a user intention to interact with the speech recognition device, wherein speech data is not buffered by the speech recognition device in the non-active mode and the user activity comprises a turn or a physical change in proximity to the speech recognition device to enable the active mode without the speech recognition device detecting the signature word;
- based at least in part on detecting the user activity, causing output of a prompt requesting user consent for the speech recognition device to collect and process speech data in the active mode by generating for display:
- a user-selectable box that is selectable to provide the user consent;
- and
- privacy terms of the speech recognition device to be consented to;
- and
- in response to receiving an input to select the user-selectable box for providing the user consent, and without the speech recognition device detecting the signature word, enabling the active mode and buffering the speech data based on an audio signal captured at the speech recognition device operating in the active mode, wherein the speech data is buffered in the active mode irrespective of whether the speech data comprises the signature word;
- and
- processing the buffered speech data to detect a command or a query for the speech recognition device.
A system, comprising:
- a memory;
- and
- control circuitry communicatively coupled to the memory and configured to:
- detect, while a speech recognition device is in a non-active mode and a signature word is not detected to enable an active mode, a user activity that indicates a user intention to interact with the speech recognition device, wherein speech data is not buffered by the speech recognition device in the non-active mode and the user activity comprises a turn or a physical change in proximity to the speech recognition device to enable the active mode without the speech recognition device detecting the signature word;
- based at least in part on detecting the user activity, cause output of a prompt requesting user consent for the speech recognition device to collect and process speech data in the active mode by generating for display:
- a user-selectable box that is selectable to provide the user consent;
- and
- privacy terms of the speech recognition device to be consented to;
- and
- in response to receiving an input to select the user-selectable box for providing the user consent, and without the speech recognition device detecting the signature word, enable the active mode and buffer, in the memory, the speech data based on an audio signal captured at the speech recognition device operating in the active mode, wherein the speech data is buffered in the active mode irrespective of whether the speech data comprises the signature word;
- and
- process the buffered speech data to detect a command or a query for the speech recognition device.
Dependencies are inferred from the claim text.